catalog.yml

# datannur catalog configuration
# Docs: https://docs.datannur.com/builder/

# Where the catalog web app is generated (published to GitHub Pages by CI)
app_path: ./catalog

# Fail the CI build if a dataset fails to scan or a metadata file is
# invalid, instead of silently publishing a truncated catalog
on_scan_error: fail
on_metadata_error: fail

add:
  # Every file dropped in datasets/ is scanned automatically:
  # variables, types, stats, preview
  - folder: ./datasets

  # Public URLs are scanned the same way — no file in the repo, no download
  # to maintain. Describe each dataset right here:
  - dataset: https://d37ci6vzurychx.cloudfront.net/trip-data/green_tripdata_2025-01.parquet
    metadata:
      name: NYC green taxi trips (January 2025)
      description: >
        Trip records straight from the NYC Taxi & Limousine Commission —
        48,000 trips scanned from a public URL. Variables, stats, preview
        and update date are extracted automatically; this name and
        description are the only manual part.
      link: https://www.nyc.gov/site/tlc/about/tlc-trip-record-data.page

  - dataset: https://d37ci6vzurychx.cloudfront.net/misc/taxi_zones.zip
    metadata:
      name: NYC taxi zones
      description: >
        Taxi zone boundaries matching the trip records above — a zipped
        Shapefile scanned from a public URL, with a map preview in the
        catalog.
      link: https://www.nyc.gov/site/tlc/about/tlc-trip-record-data.page

# Optional: enrich scanned datasets or reference external ones with
# CSV/Excel metadata files — see "Editing metadata" in the README.
# metadata_path: ./metadata